Early Fur-trade Forts of the Peace River Area of British Columbia

Early Fur-trade Forts of the Peace River Area
of British Columbia
K. R. FLADMARK
In the late eighteenth and early nineteenth centuries, the area now known
as British Columbia was being approached and penetrated from both
land and sea by Europeans intent on trading for furs with the natives of
one of the last "untouched" regions of North America. The early maritime fur trade has long been understood as a major influence on the postcontact characteristics of Northwest Coast Indian cultures, and as a result
intensively studied by ethnohistorians and "pure" historians alike. Surprisingly, however, relatively little attention has been devoted to an equivalent
understanding of the processes and evolution of the land-based fur trade
as it unfolded in British Columbia specifically and as it influenced the
interior native peoples.
It is my intention simply to describe a body of archival and archaeological data from a portion of the interior of B.C. which, if properly
exploited, has the potential to provide significant insights into many of
these and other problems. These data concern the early ( 1793-1823) furtrade period in the Peace River valley of northeastern British Columbia.
Lying east of the Rocky Mountain divide, the Peace River district often
is neglected in modern southwesterly dominated perceptions of this province, but during the late eighteenth and early nineteenth centuries it was
at the leading edge of European land-based expansion in Canada and
was the main east-west communication route by which new materials and
ideas penetrated the intermountain interior.
In the remainder of this paper I hope briefly to outline the evolution
of the fur trade physical establishment — i.e., forts and trading posts —
in the Peace River valley, based on published and unpublished archival
data and archaeology. I am indebted to Finola Finlay (1976, 1978, n.d.)
for much of the reconstruction of the fur trade history of the Peace River
area which follows. Preliminary archaeological investigations conducted
on these sites in 1974, 1975 and 1976, in the course of resource impact
assessments carried out in proposed Hydro reservoirs, also serve to support
and complement the documentary record.
48
BG STUDIES, no. 65, Spring 1985
Early Fur-trade Forts of the Peace River Area
Fur-trade Forts of the Upper Peace River Valley,
49
ijg4-i823
Sir Alexander Mackenzie, working for the Northwest Company
( N W C ) , was the first European known to have reached the Peace River
area of British Columbia, on his famous trip to the Pacific in 1793.
Prophetically, passing the mouth of what is now the Pine River, near the
present location of the city of Fort St. John, Mackenzie commented on
the suitability of that area for a future trading establishment (Mackenzie, 1793).
In 1797 John Finlay retraced Mackenzie's voyage up the Peace River
and explored the northerly branch now named after him. Unfortunately,
Finlay's journals are long lost and we know few details of his stay on the
Upper Peace. However, later secondary references suggest the possibility
that Finlay built the first fort in the area as early as 1794, near the location suggested by Mackenzie (Finlay, 1976). Whatever its initial date
there is no doubt that a N W C post, known as "Rocky Mountain Fort,"
was in operation near present day Fort St. John by, at least, 1798 (fig. 1 ).
The existence of such an early permanent European establishment on
mainland British Columbia has not always been clearly recognized by
historians (e.g., Orrnsby, 1958) and some confusion has surrounded its
precise location (e.g., O'Neil, 1928). However, substantial archival data
and archaeological verification now leave no valid basis to doubt its
existence or age.
A relatively detailed anonymous journal describes life at this lonely
establishment in the winter of 1799-1800, when it was the most westerly
English- (and French-) speaking outpost on the continent of North
America (O'Neil, 1928). The same journal indicates that the fort had
been in existence for at least one year previously, and the need to refurbish buildings which had fallen into disprepair suggests an even greater
age. The fort establishment included at least a "shop" (and/or a bighouse), men's houses (for twelve men, four women and five children), at
least one hangard (or storage cellar), a 55' flagpole and a fur-press, as
well as at least five Indian lodges nearby. We know that they obtained
their drinking water from the river, as well as throwing some of their
refuse on the beach, and the tensions of a hard, isolated existence are
revealed in terse descriptions of a French-Canadian workman who went
deleriously insane, a fight between two other men, and the firing of a shot
through the window of the chief trader's bedroom by overzealous revellers
on New Year's day. Rocky Mountain Fort continued in existence until at
least the fall of 1804, and perhaps as late as the summer of 1805. It was
50
BG STUDIES
visited in early March of 1804 by David Thompson, who undertook a
solitary and apparently miserable walk up the ice of the Peace River from
Fort Fork, a distance of about 250 km. Thompson's manuscript course
descriptions, latitudinal calculation and later published map were instrumental to re-location of the old fort site in 1975. His terse comment after
thirteen days' gruelling march, " 2 miles to the House Thank God," says
much about the realities of early fur-trade travel.
In 1805 James MacDougall built "Rocky Mountain Portage House"
on the south bank of the Peace River, opposite the present town of
Hudson Hope, and Rocky Mountain Fort was abandoned. Journals kept
by John Stuart and Simon Fraser in 1805 a n c ^ J 8o6 describe life at this
fort, and the difficulties involved in persuading the Indians to come to the
new post, which was mainly built to facilitate expansion of trade westward into New Caledonia. Rocky Mountain Portage House was situated
at the eastern end of the Rocky Mountain Portage trail, which skirted the
Peace River canyon and provided the main overland route of communication into intermontane British Columbia until the growth of the Saskatchewan and Columbia River-based supply systems in the 1820s. Rocky
Mountain Portage House was abandoned in 1814 and reopened briefly
in 1823-24 by the infamous Samuel Black under Hudson's Bay Company
(HBC) control. It was finally re-established on the north bank of the
river, at the present site of Hudson Hope, about 1866, as a supply centre
for the Omineca gold rush (Finlay, 1976).
The next post to be established along British Columbia's Peace River
valley was called "St. John's" by the Northwest Company, which owned
it, and "Fort d'Epinette" or "de Pinette" by the Hudson's Bay Company,
which eventually arrived as competition. This fort was constructed on the
north bank of the Peace River just downstream from the present Beatton
River confluence in 1806 (Wallace, 1929), and was under the control of
Fred Goedike, J. Clarke and Archibald McGillivray between 1808 and
1813, as noted by Daniel Harmon in his travels between Dunvegan (now
in Alberta) and New Caledonia (Harmon, 1800-16). There are no
known references to St. John's between 1813 and 1820, but it is presumed that it was continuously occupied through this period, as part of a
Northwest Company monopoly in the Upper Peace River and New
Caledonia. By 1820 the Hudson's Bay Company began to penetrate into
northern British Columbia and after 1821 took over control of N W C
assets.
Before 1821 there is little useful information about the physical appearance of St. John's; however, an unpublished manuscript journal kept at
Early Fur-trade Forts of the Peace River Area
51
the post by Hugh Faries between October 1822 and May 1823 contains
much detail about its installations, personnel and activities in its final
phase of existence. We learn that Mr. Faries was dissatisfied with the
poor condition of the fort, describing it as "all fallen in ruins, the wood of
the buildings being perfectly rotten," and he put his men hard to work in
repair, including construction of a new floor and chimney in one house,
putting in windows, and "mudding" all buildings with a "yellow earth"
obtained upstream along the Peace River. Considerable excitement surrounded the arrival of the first horse to reach northern British Columbia
in October 1822, obtained from Dun vegan after a failed attempt the
previous year, and men were put to work building a stable and cutting
grass. Also notable are recurrent complaints that game was becoming
increasingly hard to find in the vicinity of the fort and that some of the
Indians were near starvation. Much of the meat that was obtained was
sent westward as pemmican to supply the posts in New Caleodnia. Considerable efforts were expended in cutting birch bark and firewood, each
man being expected to cut and haul fifteen cords in preparation for the
winter. Birch trees, whose bark was so essential for the company canoes,
do occur in useful quantities and size along the Peace River, and this area
provided birch bark for New Caledonia. Interestingly, an Indian, "Chimarouche," first met in the Rocky Mountain Fort journal of 1799-1800,
is encountered again at St. John's on 14 January 1823, when, old and
crippled, he met a gruesome death by rolling accidentally into his campfire. The same character, sometimes called "Jimathush," undoubtedly of
some distinction during the early fur-trade period, but now locally forgotten, still gives his name to at least one prominent landmark in the area
today — "Jim Rose" hill, a high cliff on the north bank of the Peace
River, upstream from Rocky Mountain Fort (Butler, 1873; MacGregor,
!952).
On 14 May 1823 Mr. Faries travelled downstream to Fort Dunvegan,
while Hughes was left on the upper Peace to oversee the company's
planned closure of St. John's and the transfer of its trade to Dunvegan
and the re-opened Rocky Mountain Portage House. On the 5th of August
the Dunvegan journal (Mcintosh, 1822-23) n °tes the arrival of a message from St. John's claiming that the Indians there were "ungovernable
and in a state of open warfare." O n the 28th of October Guy Hughes
was left alone at St. John's, after sending his men upstream with goods
being transferred to Rocky Mountain Portage House. Five days later he
was shot and killed from ambush while walking on the riverbank, and the
next day four of his men, returning from upriver, were also killed as they
52
BG STUDIES
landed their canoe in front of the fort. The Indian perpetrators of the
"Fort St. John massacre" looted the buildings but did not set fire to them,
and were just disappearing across the river when another HBC canoe
happened to arrive downstream from Rocky Mountain Portage House
(Heron, i824;Krech 5 1982, 1983).
The company launched an investigation of this event, but the "bandittes" were never captured. A wide range of reasons has been cited to
explain the massacre, including the opinion, still stated by living Indians
in the area today, that Hughes was thought to have magically or supernaturally injured a native boy (Goddard, 1916; Ridington, 1979). Whatever Hughes' magical capabilities, the basic reasons for the attack were
probably economic, centred on the plan to close St. John's as a £ 'restraint"
measure because of generally failing resources in the area, which would
have forced the "St. John Indians" to trade in territories of unfriendly
neighbouring groups. As a result of the massacre, which was one of the
most violent events ever to occur in interior British Columbia involving
fur-trade company personnel, all HBC posts on the Upper Peace River
were closed and none re-opened in the British Columbia portion until
about i860.
The 1823 massacre at St. John's should, seemingly, close the book
on the early fur-trade period in the upper Peace region. Later reestablishments of Fort St. John (Fort St. John II, ca. 1860-74; Fort St.
John I I I , 1874-ca. 1923) closer to the present city which bears that
name are beyond the period considered in this paper, as is Rocky Mountain Portage House of the post-1860 period. However, there is another
brief but complicating chapter which requires mention, and it takes us
back to the site of old Rocky Mountain Fort.
Between about 1818 and 1820, the Northwest and Hudson's Bay
companies were engaged in intense and sometimes bitter competition for
diminishing fur resources. Some of the most extreme episodes in this
struggle occurred in the Athabaska District, which included the Peace
River area. In 1820 the HBC established its first fort in British Columbia,
in the Peace River valley, at the site of old Rocky Mountain Fort about a
half day's canoe travel above St. John's. This new post, variously known
as "Fort de Pinette," "Yale's House" and "Fort George," was intended
to provide direct competition for the NWC at Fort St. John's and to be a
springboard for expansion into New Caledonia (e.g., Simpson, 1938). To
further complicate matters, the NWC, in turn, set up a nuisance guardpost directly beside the new HBC establishment as was the pattern in
earlier encounters between the two trading rivals. Thus, briefly in 1820-
Early Fur-trade Forts of the Peace River Area
53
2 i , there were three separate forts operating simultaneously in a single
short stretch of river valley. The fact that at least two of these forts shared
the same name ("Fort de Pinette"), while two were also on the site of a
less well-known and much older fort, has considerably complicated interpretation of the historical record.
In 1821 the Hudson's Bay and Northwest Company merged, under the
energetic and parsimonious direction of Governor George Simpson. One
of his major concerns was rationalization of fort sites and termination of
redundant establishments constructed during the period of most intense
competition. As part of this campaign, "Yale's House" and its NWC
shadow at the old site of Rocky Mountain Fort were closed, company
operations shifted to the old NWC fort of "St. John's," and an unsuccessful attempt was made to float the buildings from "Yale's House" downstream to "St. John's." Thus, in the final two years of its seventeen-year
history, that establishment operated under new company colours, to the
confusion of some secondary historical chroniclers (e.g., Bowes, 1963;
MacGregor, 1952; Wallace 1929), not to mention even later Hudson's
Bay Company personnel (e.g., Alexander Mackenzie, 1908), concerning
the age, locations and historical associations of the various predecessors of
the modern town of Fort St. John.
Fur-trade Archaeology in the Upper Peace River Area
In the summers of 1974, 1975 and 1976 I directed aspects of archaeological surveying and excavation in the proposed " C " and " E " hydroelectric reservoirs in the Peace River valley, between the Alberta border
and Hudson Hope, funded by B.C. Hydro through what was then the
Provincial Archaeologist's office. Although the sheer numeric density of
prehistoric aboriginal sites in the area consumed most archaeological
efforts, a concern was also the re-location and identification of the historically documented fur-trade forts described above.
The site of Rocky Mountain Portage House was archaeologically
relocated in 1974, readily visible as a complex of low mounds and shallow
depressions enclosing a roughly quadrangular area. Unlike other fur trade
sites in the Peace River valley, this establishment has not been damaged
by riverbank erosion, but has been affected by modern construction
activities, including several relatively recent log-cabins and a road and
landing area for a Department of Highways ferry which operated here
until completion of the present highway bridge about 1965. Other than
basic mapping and surveying, no archaeology has yet been undertaken at
54
BC STUDIES
Rocky Mountain House, since it would have been least affected of all the
early fort sites by proposed Hydro dam construction.
In 1975, armed with primary historical documentation, such as
Thompson's journal, we were able immediately to locate the site of
Rocky Mountain Fort. In addition that year, we began excavations at the
Beatton River site with the SFU archaeology field school. Results of that
work, and archival data obtained that winter, permitted increasing confidence in identifying the Beatton River site as "St. John's" of the 1806-23
period, including both N W C and later HBC occupations. The definite
presence of the HBG at this site was further supported by a second season
of excavations in 1976, again with the SFU field school, which saw the
recovery of several indisputably HBG artifacts, including a company seal.
That season also saw some small exploratory excavations at Rocky Mountain Fort, under the direction of Finola Finlay (Fladmark, 1975; Fladmark, Finlay and Spurling, 1977; Spurling, Finlay and Fladmark, 1976).
As first observed in 1975, historical remains at Rocky Mountain Fort
consisted of three spatially distinct groups of surface features, spread
along more than 180 m of riverbank on a low floodplain terrace of the
south bank of the river (fig. 2 ) . The most westerly or furthest upstream
group ("Area A") consisted of two sandstone fireplaces eroding from the
bank, and some low mounds and pits. Test excavations directed by Finola
Finlay in 1976 revealed the outline of a single house, approximately
6 x 9 m in area, containing very few artifacts (one lead musket ball, one
piece of sheet glass, and a plain brass button). The nature of these artifacts indicates a relatively early contact age, but their scarcity—in
marked contrast to other portions of this site as well as St. John's —
strongly suggests a brief occupation. Most of Area A has probably been
lost to river erosion, but still there is evidence to suggest that it represents
one of the 1820-21 occupations — perhaps the N W C nuisance post —
rather than Rocky Mountain Fort itself, which was occupied for six to
ten years.
Approximately 100 m downstream from "Area A," a well-defined,
narrow trench encloses a rectangular area approximately 26 x 30 m in
size, with its long axis at right angles to the riverbank. No obvious structural features occur within the enclosure, although a narrow test trench
revealed possible burned beams, as well as gun-flint fragments, a red trade
bead, a piece of retouched chert, a fragment of sheet glass and some birch
bark. This feature ("Area B") is considered to be the remains of a
stockade trench — one that perhaps was never filled, or from which the
posts were later removed — again resulting from one of the 1820-21
Early Fur-trade Forts of the Peace River Area
55
occupations, possibly that of Hudson's Bay Company "Yale's House."
The latter suggestion is reinforced by the information in Hugh Faries'
journal that an attempt was made to remove those structures and float
them downriver to St. John's.
Centred about 30 m downstream beyond the east edge of this stockade
trench, "Area C" is the most complex of the three feature clusters
(Spurling, Finlay and Fladmark, 1976). It includes at least three stone
chimney mounds, three sizeable'"cellar" depressions, and several smaller
mounds and hollows, in an area approximately 30 x 15 m in size, paralleling the river bank. A magnetometer survey conducted by Dr. John
Stroud, then a visiting physicist at SFU, clearly revealed a ca. 8-9 m long
rectangular wall-outline, including a possible door, enclosing the two
largest chimneys, as well as possible suggestions of a stockade line, not
otherwise visible (Sparling, Finlay and Fladmark, 1976). Test excavations
around the periphery produced glass beads, gun-flints, a small brooch, an
embossed steel button, a clasp-knife blade, part of a mattock-head ( ? ),
an iron arrowhead, file fragments, a gun lock-screw, lead shot, nails,
fragments of sheet copper, and a fragment of an incomplete stone pipe
bowl of so-called "Micmac" type. The density of features and artifacts in
"Area C" suggests a substantial and relatively long-lived occupation.
Although excavations conducted to date have been simply exploratory,
there is reason to believe that this area represents the remains of old
Rocky Mountain Fort itself. If that is the case, then this site is remarkably well-preserved and completely unaffected so far by modern disturbances of any sort, making it probably unique among western Canadian
fur trade forts of this age. It is possible, however, that some part of the
front of the fort has been removed by riverbank erosion; that won't be
determined until full excavations can be carried out.
It seems too good to be true that a site for which there is documentation of three distinct historical occupations should produce relatively clear,
unmixed surficial archaeological evidence to support this, let alone that
the oldest of these occupations should turn out to be the best preserved !
It was with considerable relief that I greeted the recent news that B.C.
Hydro has decided to forgo, for the time being, any further development
of the " C " damsite on the Peace River. Not only would that project have
deeply submerged Rocky Mountain Fort, but the old fort itself lies sufficiently close to the actual damsite that it would undoubtedly have been
directly and seriously damaged by construction activities. Now we still
have the opportunity to ensure that it be properly investigated and pro-
56
BG STUDIES
tected — there are not many equivalent site complexes remaining anywhere in western Canada, and for British Columbia it is unique.
St. John's, at the mouth of the Beatton River, was not burned by the
Indians after the 1823 "massacre," and apparently the gloomy but
habitable buildings still stood as late as 1833 (McLean, 1833). Crosses
on the beach marked the deaths of Hughes and his men, and occasional
passersby would camp briefly in the mouldering old fort (McDonald,
1828; McLean, 1833). By the 1870s, however, St. John's was reduced to
mere overgrown chimney mounds, although the drama earlier enacted
there was not forgotten (Butler, 1873). Even .modern residents are still
aware of parts of the story, and the old fort site has never been completely lost. In 1974, acting on secondary historical sources and the advice
of local landowners, we readily relocated the remains of "St. John's"
amid the dense cottonwood and willow tangles of the low river floodplain
terrace which has held them for about 150 years. In 1975 and 1976,
supported by B.C. Hydro, SFU archaeological field schools intensively
excavated this site under my direction, and some results of this work will
be described in a following section.
In comparison to the very limited investigations carried out at Rocky
Mountain Fort, our excavations at St. John's in 1975 and 1976 were
relatively large-scale. Over 100 1.7 x 1.7 m pits were dug, resulting in the
recovery of approximately 5,000 artifacts, thousands of animal bones, and
large amounts of structural information. There is not space here to
describe all of this in any detail, so the following represents general reconstruction of the fort and some of its activities based on many lines of
evidence (Spurling, Finlay and Fladmark, 1976).
St. John's was a stockaded fort, with the outer wall originally enclosing
an area of approximately 30 x 33 m. The precise dimensions cannot be
known with certainty, since an unknown portion of the front of the site
was removed by river erosion sometime after its abandonment. The
stockade itself was relatively light and flimsy, in strong contrast to the
massive fortifications of both prairie and coastal forts, probably attesting
to the normally peaceful conditions expected in this area. Within the
stockaded area there were three main buildings, enclosing a central quadrangle or "parade-ground," probably originally about 14 m square (fig.
3 ) . On the north side of the quadrangle was the "big-house," which
served as chief trader's residence, store and general mess-hall. Measuring
approximately 7 x 1 2 m , the ground plan of this structure indicates three
main rooms. Occupying the western two-thirds of the house was a single
large (dining?) room overlying a deep cellar, heated by a sandstone
Early Fur-trade Forts of the Peace River Area
57
fireplace diagonally oriented in one corner The remaining eastern portion
of the structure was divided into two chambers, with a partition largely
formed by a sandstone double fireplace and chimney, ca. 4 m high, one
hearth facing into each room. At least one of these rooms was probably
the clerk's bedroom, as suggested by a straight razor and other personal
items. Archaeological data do not provide much information about the
architectural details of this house above the floor, although it was likely
of post-on-sill construction, with a bark-covered roof. It was floored with
well-finished planks, fitted with tongue-in-groove joints, and its walls and
chimneys had been repeatedly "washed" with yellow earthy pigment. It
is likely that the eastern portion of the house represents a newer addition,
and at least one of the fireplaces and some of the floor had been completely rebuilt in the course of its history. The cellar contained gun-parts
and a bronze barrel spigot, suggesting its use for storage of valuable or
sensitive materials such as weapons and liquor. Clay mounds accumulated
around each fireplace and protecting the floor planks indicate that the
roof had deteriorated sufficiently to permit the mudding to slough off the
chimneys before the house met an ultimate fate through fire. Considerable quantities of nails found in the west fireplace suggest that nailed
lumber was being burned, probably by casual users of the building after
its abandonment.
The other two buildings are far less well preserved, and were possibly
much less substantial structures to begin with. On the west side of the
quadrangle lies the remaining portion of the "men's house," so identified
by the quantity and diverse ethnic identiy of artifacts found within, and
the known usual placement of such a barracks building along one side of
a fort's central area. This structure had an earthen floor and was probably divided into two or three individual rooms adjoining in a row, each
with its own single hearth and chimney. Quantities of native-manufactured artifacts were found in this building on the floor, sunk into it, and
deep beneath in several sub-floor garbage pits. Such artifacts include
stone tobacco pipes of both plain-elbow and "Micmac" type in various
stages of completeness, including one very large specimen ornately decorated with a lead inlay; bone snowshoe needles and awls; a toothed
metapolial-bone flesher; occasional flaked chert and quartzite artifacts;
and birchbark items, including remains of decorated baskets, geometric
bark cut-outs and quantities of cut birchbark scrap. These were in
addition to objects of Euro-Canadian manufacture, including various
clasp-knives, "canoe-awls," burning glasses, gun parts and balls, beads,
58
BG STUDIES
buttons, clay-pipes, ornaments, and even a complete leather shoe, fortuitously preserved in alkaline wood-ash deposits in a sub-floor pit.
Opposite the men's house, on the east side of the quadrangle, the
presence of a probably comparable building is just demonstrated by a
small remaining corner which has not yet eroded onto the river beach.
This building had a plank floor over a stone "foundation," and the
presence of a number of small glass beads and equivalently sized lead
bird-shot filtered down between the floor planks suggests the possibility
that it originally functioned as a storehouse or trading store.
Behind the big-house a deep cellar, lacking evidence of any substantial
overlying structure, may represent the remains of a "hangard" (a semisubterranean storehouse for meat) or perhaps a saw-pit. Just outside the
back wall of the stockade, a low rambling midden mound consists of
stratified ash and humic layers, as well as numerous artifacts and bones,
indicating that there was nothing complicated about refuse disposal
patterns at this site. Other pit features both inside and outside the stockade line suggest the presence of further structures not yet revealed. In
summary, the final form of St. John's may have approximated the artist's
reconstruction shown in figure 4.
The archaeology of St. John's provided direct confirmation of many
details of the historical accounts of the site, or conversely might have
permitted reconstruction of many details in the absence of written descriptions. Areas of compliance between the documentary and archaeological
records include site location; evidence of both Hudson's Bay Company
and Northwest Company occupations; evidence of building repair and
reconstruction; mudding of buildings with yellow earth; and evidence of
a considerable period of slow decay prior to burning. Even the famous
horse, so proudly brought to the area in 1822, may be archaeologically
represented by a butchered horse mandible, the only domestic animal in
the large faunal assemblage from the site, other than dog (Williams,
1978). Perhaps the horse was killed by the Indians at the time of Guy
Hughes' murder, or died earlier from any number of possible natural
causes, to be eaten by its one-time owners. In contrast, a feature definitely
present archaeologically, for which there is a puzzling total lack of comment in the documents, is the outer fence or stockade. Perhaps it predates
the 1822-23 (HBC) period for which we have most historical data, or
was considered totally insignificant by Faries and the other chroniclers.
In conclusion, a large body of documentary and archaeological data
bears on the history of the early fur trade period in the Peace River
Valley of British Columbia. Many of these data remain unanalyzed and
Early Fur-trade Forts of the Peace River Area
59
unintegrated, but even in their present preliminary form they establish
the basic site chronology and provide bases for mutual verification of
archaeological and documentary evidence. The full depths of both documentary and archaeological records have barely been touched as they
pertain to subsistence and settlement patterns, identification of indigenous
and extra-areal ethnic groups (e.g., Crée, Iroquois, Métis, Highlanders,
French Canadians), establishment of artifact seriations and numerous
other research topics. It is hoped that this paper has revealed a little of
this potential, as well as a glimpse of the rich and fascinating fur trade
history of the Peace River area.
A cknowledgements
Funding for archaeological field work in the Peace River valley in
1974-76 was provided mainly by B.C. Hydro through the Provincial
Archaeologist's Office, Victoria. Costs of excavations at St. John's by
SFU field schools in 1975 and 1976 were borne mainly by the university,
while test excavations at Rocky Mountain Fort in 1976 were partially
funded by a provincial government student employment grant. Many
students and colleagues were involved with various aspects of archival
and archaeological research. I would like especially to acknowledge Finola
Finlay's invaluable investigations in the Hudson's Bay Company archives,
Winnipeg, and her unpublished syntheses of Peace River fur trade history
on which much of this paper is based. In addition I would like to thank
Shirlee Anne Smith, chief archivist of the Hudson's Bay Company
archives in the Provincial Archives of Manitoba, for her assistance. David
Burley, Stan Copp, Finola Finlay and Ian Wilson were teaching assistants
during excavations at St. John's, while Drs. David Huntley and John
Stroud (Physics, SFU) carried out magnetometer and metal-detector
surveys of the sites of St. John's and Rocky Mountain Fort. Finally,
special thanks are due Mr. and Mrs. Richard Hadland of Dawson Creek,
B.C., who let us camp on their property in 1975 a n c ^ x 97^References
Bowes, Gordon E. (editor)
J
9^3 Peace River Chronicles, Prescott Publishing Co., Vancouver.
Butler, William F. (1873)
1873 The Wild North Land, Hurtig Publishers, Edmonton (1973).
Faries, Hugh
1822- Journal of Occurances. St. Johns. 1822.3.
23 Ms. Journal, Hudson's Bay Co. archives B.i89/a/i.
60
BC STUDIES
Finlay, Finola
1976 A summary of fur-trade history in the Peace River District of British Columbia—1794-1824. Unpublished m s - j D e p a r t m e n t of
Archaeology, SFU.
1978 Historic archaeological resources. I n : The Site C Heritage
Resource Inventory and Assessment, B. Spurling, editor. Report submitted to B.C. Hydro, pp. 45-52.
n.d.
History and mythology: T h e murders at Fort St. John, U n p u b lished ms., Charlie Lake, B.C.
Fladmark, K. R.
1975 Peace past: A report on the archaeological reconnaissance of the
Peace River basin, 1974 field season. Unpublished report submitted
to the Archaeological Sites Advisory Board and B.C. Hydro.
Fladmark, K. R., F. Finlay, and B. Spurling
1977 Archaeological investigations in the Peace River basin of British
Columbia: Summer 1975. I n : Annual Report for the year 1975.
Activities of the Archaeological Sites Advisory Board of British
Columbia and Selected Research Reports, B. Simonsen, editor.
Ministry of Recreation and Conservation, Victoria, p p . 276-320.
Fraser, Simon
1806 Journal at Rocky Mountain Portage. I n : Simon Fraser, Letters
and Journals, 1806-1808, W. K. Lamb, editor. Macmillan Co.,
Toronto, pp. 163-84. (i960)
Goddard, E. P.
1918 T h e Beaver Indians. Anthropological Papers of the American
seum of Natural History, 1 0 ( 4 ) , pp. 204-92.
Harmon, Daniel W.
1800- Sixteen Years in the Indian
16 Macmillan Co., Toronto.
Country,
Mu-
W. K. Lamb, editor (1957).
Heron, Francis
1824 A statement of circumstances connected with the murder of M r .
Guy Hughes a n d four men at St. John's in Peace River, by the
Indians of that place on the 2nd and 3rd of November last. M a n u script D 4 / 8 7 , Hudson's Bay Company Archives, Provincial Archives of Manitoba, Winnipeg.
Krech, Shepard
1982 T h e banditte of St. John's. Beaver 313:3, pp. 36-41.
1983 T h e Beaver Indians and the hostilities at Fort St. John's.
Anthropology 20(2) : 35-46.
MacGregor, James G.
1952 The land of Twelve-Foot
Edmonton.
Arctic
Davis. T h e Institute of Applied Art, Ltd.,
Early Fur-trade Forts of the Peace River
61
Area
Mackenzie, Alexander
1908 (description of early Fort St. J o h n ) Ms. on Peace River District,
E d m o n t o n Public Archives Ref. 76.377.
Mackenzie, (Sir) Alexander
1793 First Man West. Alexander Mackenzie's journal of his Voyage to
the Pacific Coast of Canada in 1793, W. Sheppe, editor. University
of California Press ( 1962).
McDonald, Archibald
1828 Peace River. A canoe voyage from the Hudson's
H u r t i g Publishers, Edmonton ( 1971 ) .
Bay to Pacific.
Mcintosh, William
1822- Journal of the transactions of the post of Dunvegan. Ms. journal,
23 Hudson's Bay Company Archives B . 5 6 / a / i .
McLean, John
1833 (Peace River journey from Dunvegan to New Caledonia) I n :
Notes of a Twenty-five
Year's Service in the Hudson's Bay Territory, W. S. Wallace, editor. Champlain Society, Toronto (1932).
O'Neil, Marion (editor)
1928 Journal of the Rocky Mountain Fort, Fall 1799. Washington
torical Quarterly 1 9 ( 4 ) , pp. 250-70.
Ormsby, M a r g a r e t A.
1958 British Columbia:
His-
A history. Macmillan Co., Toronto (1971).
Ridington, Robin
1979 Changes of M i n d : Dunne-za resistance to empire. BC Studies 4 3 :
65-80.
Simpson, Sir George
1938 Journal of occurrences in the Athabasca Department,
1820 and
1821, and report, E. E. Rich, editor. Hudson's Bay Records Society,
Toronto.
Spurling, Brian E., F. Finlay and K. R. Fladmark
1976 Report on the Peace River archaeological survey and salvage project, 1976 field season. Unpublished report submitted to the Archaeology Sites Advisory Board, Victoria, and B. C. Hydro.
Stuart, J o h n
1805- Journal at the Rocky Mountains, December 20, 1805. Ms. journal,
06 Bancroft Collection, Pacific Coast ms., University of California,
Berkeley, Series B, No. 19. (Another copy is in the Public Archives
of C a n a d a , O t t a w a — M . 731.)
Thompson, David
1804 Journey from the Forks of the Peace River to the Rocky Mountain
House 1804 — N W Coy. Ms. journal in Ontario Public Archives,
Toronto. Microfilm copy at U B C library.
1
62
Wallace, J. N.
1929 The wintering
BG STUDIES
partners on Peace River. Ottawa.
Williams, Jean H .
1978 Fort D'Epinette: A description of faunal remains from an early fur
trade site in northern British Columbia. Unpublished M A thesis,
Department of Archaeology, SFU.
FIGURE 1
Fur trade forts of the upper Peace River: 1. Fort Fork (NWG, established
1792); 2. Rocky Mountain Fort (NWG, 1794?), Fort d e Pinette/Yale's
House (HBG, 1820), Mcintosh's "nuisance post" (NWG, 1820); 3. Rocky
Mountain Portage House (NWG, 1804) ; 4. Dunvegan (NWG, 1805) ; 5. St.
John's (Fort St. J o h n I, NWG, 1806) ; 6. Fort St. John (II, HBG, i860) ; 7.
Rocky Mountain Portage House/Hudson Hope (HBG, 1866) ; 8. Fort St.
John ( I I I , HBG, 1874). Not all Alberta forts shown.
Early Fur-trade Forts of the Peace River Area
63
24 m
t=
\ • I
=/
20 cm contours
"0
CD
Q
O
CD
DC)
<"
CD
FIGURE 2
Preliminary map of the Rocky Mountain Fort site. Areas "A", "B" and "C"
are clusters of surface features which may correspond respectively to the
1820 NWG "nuisance post," Yale's House and Rocky Mountain Fort.
Hachured ovals represent cellars and solid ovals are stone chimney mounds.
The dotted area paralleling the river bank represents a gravel bar which
may provide some indication of the extent of shoreline erosion since the forts
were occupied.
64
BG STUDIES
FIGURE 3
M a p of excavated features, St. John's 1806-1823. a. stockade line; b. "bighouse"; c. possible dining room-mess hall; d. cellar; e. corner fireplace; f.
double fireplace separating two rooms in the eastern portion of the bighouse; g. "men's house"; h. fireplace in men's house; i. sub-floor pits; j .
"store-house"; k. "hangard" or saw-pit; 1. garbage midden; m. internal
fence-lines; n. possible gate; o. rock-filled pit; p. other pits. Black areas are
charred planks and timbers.
Early Fur-trade Forts of the Peace River Area
65
FIGURE 4
Artist's impression of the possible appearance of St. John's in 1823. The
dashed line indicates the position of the modern riverbank.